Datasets:

ShuoShuoShuo

/

PromptsLabels

like 0

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A blue iguana space mechanic wears spacesuit armor with cybernetics.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "animals, people, vehicles" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A perfume bottle sits on a small mountain of golden sand.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "landscape,fashion" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Elderly woman screams at Jesus.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A gothic vampire and necro witch summon a cosmic god of torment at an altar.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "NSFW" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Aesthetic girl holds a stunning stained glass sword.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, portrait, art" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A tall, slim woman dressed in red leather challenges Satan.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,fashion,portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A beautiful glowing woman walks over calm waters with a massive ark in the background.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, landscape" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Arthur Morgan wears a catsuit as a cyberpunk assassin.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,fashion,portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A highly detailed portrait of a dark witch with red glowing eyes.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "portrait, people" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A male werewolf devours a human arm in a dark forest, surrounded by blood.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "NSFW" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A Chinese Taoist fairy goddess in Elden Ring attire.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,fashion,portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A pregnant black widow wears a red and black bikini.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "NSFW" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A white marble Greek goddess face profile sculpture with raw flesh underneath.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "art,portrait,people" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A flock of radiant seraphim emerges from a glowing vulva-shaped temple doorway.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "NSFW" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A beautiful anime girl with glasses and long blue hair gazes.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Time travelers from different eras appear in a park.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, landscape" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A glowing electrical Steampunk space rocket stands in a Victorian museum.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "indoor,art,vehicles" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Two teenagers talk on a rooftop overlooking Shanghai Bund.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, landscape" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Caligula lies dead on ancient amphitheater's stairs.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Emma Stone poses as a mysterious thief in dark attire and a mask.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,portrait,fashion" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [An alien dream world with surreal architecture and mega structures.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "landscape" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A cyborg Dwayne Johnson in a futuristic spacesuit stands in space.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,portrait,fashion,landscape" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [An empty and abandoned living room with a flight of stairs in the hallway at night.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "indoor" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A melanated woman holds a cyborg Batman baby, eating crayons.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,portrait,fashion,food" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A demon noblewoman with ram horns wears a elegant outfit in a close-up portrait.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "portrait, people, fashion" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A majestic white lion wizard casts a blue spell.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "animals, art, people, portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Uma Thurman, dressed in a yellow kimono, wields Hattori Hanzo's katana amidst autumn leaves.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,fashion,landscape,plants,portraits" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [This is a portrait of DC's Martian Manhunter.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "portrait, people" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A snake hunts a butterfly amidst withered flowers in a lush still life arrangement.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "animals, plants, art" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A white angel radiates a giant ball of miracle light.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A teenage Asian android pilot sits in a cybernetic cockpit.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, vehicles, indoor" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A beautiful BIPOC woman with a scar on her cheek wears FemShep Mass Effect armor.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,portrait,fashion" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A fantasy gate opens onto marble steps surrounded by flowers and ruins.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "landscape, plants" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A futuristic dieselpunk street littered with garbage and cables.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "landscape, vehicles" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A cat holds a teacup in a living room setting.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "animals, indoor" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [SpongeBob wears tribal regalia in a divine portrait.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "portrait,art,fashion,people" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A kid gazes up at a gigantic abandoned robot in bright sunlight.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "landscape, people, vehicles" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Tom Cruise transforms into a fiery alien harpy queen.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,portrait,animals" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A mermaid swims near a vibrant coral reef ecosystem.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "animals, landscape, plants" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A beautiful Greek goddess looks up into the pouring rain with soaking wet hair.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,portrait,fashion" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A futuristic cityscape with neon-lit skyscrapers and lush greenery.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "landscape, plants" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A pile of apples forms a lifelike portrait of Steve Jobs' face.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "food, portrait, people" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [An anthropomorphic rat in pajamas lies on a bed in a cluttered room.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "animals, indoor, people" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A boy with curly black hair and glasses is posing for a portrait.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "portrait, people, indoor" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A standing elf in golden and silver armor holds a curved long sword.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, portrait, art" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A portrait of a tall 40-year-old woman with striking features and black attire.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "portrait, people" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A shiba inu stands in a graveyard, wearing holy armor.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "animals, art" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Adriana Lima wears Skyrim MOD Angel Armor with wings open.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,fashion,portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A beautiful girl with red hair and freckles poses for a close-up portrait.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,portrait,fashion" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A scary masked radioactive lizard of Nazareth with giant wings.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "animals, art" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A punk woman with a mohawk stands amidst splashes of neon galaxies and rain.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, portrait, fashion" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A crazy screaming sinister voyager wears digital sunglasses.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,fashion,portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A portrait of young Sigourney Weaver from Alien.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "portrait, people" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A giant muddy pig wanders through Soho's vibrant nightlife scene.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "animals, people, landscape" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A magical girl idol with long wavy colorful hair wears a colorful dress.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,fashion,portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A Japanese cyberpunk mercenary wears a tactical headset.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,portrait,fashion" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Mila Kunis poses as a divinity paladin in a painted portrait.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, portrait, art" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A chibi cyborg knight Sega Nightmaren wears a cockscomb cap with angel wings.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,art,fashion,portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A T-Rex astronaut explores red Badlands canyons.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "animals, landscape" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A girl in a kimono offers a peach to a large bear.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "animals, art, fashion, food, people, plants" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A rotting Asian zombie jogger stands in a city park.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, landscape" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A kiwi bird dressed as a plague doctor in a detailed portrait.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "animals, art, fashion, portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A mermaid entity trips on DMT and acid.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "NSFW" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [An abandoned aquatic exhibition in a science museum with old computer screens.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "indoor" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A kitten sits on a hill surrounded by fluffy pink clouds.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "animals, landscape, plants" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A grey seer Skaven prophet holds a ceremonial staff in Warhammer Fantasy setting.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A large, looming creature wraps around a tree with sharp teeth.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "animals, landscape, plants" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Ryan Gosling is a delicious cheesecake.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "food, people" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Gal Gadot poses as a seductive and powerful fantasy queen sorceress.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, portrait, fashion" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [An androgynous assassin with demon horns and elf ears wears blue tunic and robes.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A cute girl drives a Ferrari on a simracing simulator in her personal room.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, vehicles, indoor" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Daniel Radcliffe is portrayed as Wolverine, an American stand-up comedian in a symmetrical portrait.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A group of white female androids lie scattered on the floor, body parts disconnected.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "indoor, people" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A male Egyptian god with a bull head headdress stands as a king amidst stormy clouds.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A young woman holds an umbrella while standing on a rainy city street surrounded by skyscrapers.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, landscape" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Jessica Chastain transforms into a fiery alien harpy queen.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, portrait, art" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A goddess with pink twintails and cyan eyes smiles.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, portrait, fashion" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A black metal pagan god with antlers and glowing white eyes.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [This is an isometric yellow alien plant.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "plants" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A young punk rock man with curly brown hair plays guitar in a denim vest.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,fashion,portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Harry Kane and Son Heung-min pose as buddy cops in a vibrant full-body portrait.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A close-up portrait of an elder Soviet general in an empty Stalinist hall.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "indoor,people,portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Amy Yasbeck poses in full bronze and iron armor.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, portrait, fashion" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A giant serpent-like creature wraps around a massive tree.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "animals, landscape, plants" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A girl with white hair wears a black skirt surrounded by flowers.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, portrait, fashion, plants" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A grand Egyptian palace stands by a small pool.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "landscape, architecture" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A grungy woman with a rainbow mohawk and side boob stands in torn overalls.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "NSFW" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A jellyfish fairy princess walks in the streets of Japan.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,fashion,animals,landscape" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A massive mud sculpture explodes violently inside a modern museum gallery.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "indoor,art,people" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Will Smith slaps Chris Rock in a dark hellish scene.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "NSFW" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A ruined city environment with rusted vintage cars and trucks in a low density urban zone.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "landscape, vehicles" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Black military quadcopters swarm through a futuristic prison's wide underground hallways.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "indoor,vehicles" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A transparent glass statuette of Abraham Lincoln is displayed.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "art,portrait,indoor,people" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A Disney-inspired Princess Zelda stands confidently with the Triforce in a lush landscape.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "landscape, people, portrait, plants" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A brunette fairy woman wears a pink corset and tutu.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "fashion,people,portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Kassandra from Assassin's Creed Odyssey sits for a portrait.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people,portrait,fashion" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A cute crochet astronaut travels alone on a yarn planet.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "landscape, people, art" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A delicate female angel with long silver hair and pale skin.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, portrait" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [A Hawaiian hula dancer holds a white tea cup at beach.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, landscape, fashion" } ]

[ { "from": "human", "value": "Given the following image prompt and list of labels, \nImage prompt: [Natalie Portman poses as a legendary warrior in Lord of the Rings-inspired battle armor.] \nAvailable labels: ['animals', 'art', 'fashion', 'food', 'indoor', 'landscape', 'logo', 'people', 'portrait', 'plants', 'vehicles'] \nYour tasks are as follows:\n1. Determine whether the image contains harmful content or is not safe for work (NSFW). Harmful content includes, but is not limited to, harmful logo, extremist organization or religion, violence, crime, pornography, nudity, weapons, etc. If the image is NSFW, return the label \"NSFW\" (Not Safe for Work).\n2. Identify and select the most relevant labels that describe the key elements of the image the prompt might produce. Provide only the selected labels, separated by commas.\nIf the answer for first task is Yes, return NFSW as the selected labels, otherwise return selected labels from the second task. OOutput only the selected labels, without any additional text or explanation.\nSelected labels:" }, { "from": "gpt", "value": "people, portrait" } ]